Evidence-Based Infection Control in the IntensiveCare Unit

(1)

Evidence-Based Infection Control in the Intensive Care Unit

J. HUGHES, N. TAYLOR, E. CERDA, M.A.DE LACAL

Introduction

Nosocomial infection rates in the intensive care unit (ICU) are higher than in the general hospital population, the main reason being the severity of illness of ICU patients and hence the increased susceptibility to acquiring micro-organisms related to the ICU [1]. The unique environment of the ICU makes it more likely to encourage the emergence of infections due to multi-resistant poten- tially pathogenic microorganisms (PPMs) [2]. The ICU is the area in the hospi- tal where the most severely ill patients are brought together. Practically all ICU patients, who require mechanical ventilation for >3 days receive antimicro- bials. They are also cared for by larger numbers and a wider variety of healthcare workers (HCWs), which increases the risk of transmission of PPMs.

Additionally, the often-urgent nature of critical care interventions can lead to sub-optimal practice of infection control [3]. This can lead to increased mor- bidity and mortality, become a drain on existing resources, and increase pressure on beds. However, it is also a quality and clinical governance issue [4, 5].

The three main clinical endpoints [6] for controlling infection in the ICU are to: (1) reduce high infection rates, (2) maintain low infection rates, and (3) control antibiotic resistance. Nosocomial infections are caused by micro-organisms acquired on the ICU following transmission via the hands of carers. Hence, the target of infection control is the control of transmission of often multi-resistant micro-organisms following breaches of hygiene.

Universally agreed infection control measures are basic hygiene standards, including hand hygiene, patient isolation, appropriate glove and gown usage, care of equipment, and standards for environmental cleanliness [7–9]. The role of device and antibiotic policies will be discussed in the following two Chapters 11 and 12. Although there are a number of studies on the benefits of such measures, in general, compliance with some or all of the practices is known to be sub-optimal and the evidence base is often poor or absent [10, 11]. Attempts to

(2)

address some of the issues and to formulate national evidence-based guidelines for healthcare-associated infections have been addressed by the evidence-based practice in infection control (EPIC) project 2001 [11], Agency for Healthcare Research and Quality [10], and the Center for Disease Control and Prevention (CDC) [12, 13].

Why Do We Need Evidence-Based Practice in ICU?

Evidence-based practice (EBP) is defined as the “integration of best research evidence with clinical expertise and patient values” [14]. EBP is now recognized as a means of providing more effective health care by questioning the basis of many interventions and analyzing the available original clinical research in order to substantiate current practices and to form guidelines based on expert opinion where no other evidence currently exists [15]. EBP enables healthcare professionals to be confident that their interventions are informed by a current and appropriate knowledge base. It also helps ensure that practices and guidelines can be audited and measured against agreed standards. This is essential in the current healthcare climate and is high on many political agendas, particularly in relation to clinical governance directives [5]. The concept of EBP and its application is a key feature of clinical governance.

Although many infection control policies and recommendations have been attempted based on the available evidence, there is often a paucity of studies to support practices in infection control, with some procedures still based on ritual [16, 17]. Controlled clinical trials are difficult to undertake due to the large numbers of subjects required and the multiplicity of factors involved in infection control. In some cases, the difficulties of basing infection control practice on the best available evidence is often a problem due to the ethical considera- tions of conducting a randomized clinical trial. In addition, there are some areas of practice, such as hand decontamination, where there is some evidence to show it can help reduce infection, but compliance is often poor [18].

The National Health Service Executive in the United Kingdom [19] stated that all clinical guidelines should be classified according to: randomized controlled trials, other robust experimental or observational studies, or more limited evidence usually based on expert opinion and endorsed by respected authorities.

The CDC guidelines [12, 13] based their evidence on:

Category IA. Strongly recommended for implementation and supported by well-designed, experimental, clinical, or epidemiological studies

Category IB. Strongly recommended for implementation and supported by certain experimental, clinical, or epidemiological studies and a strong theoretical rationale

(3)

Category II. Suggested for implementation and supported by suggestive clinical or epidemiological studies or a theoretical rationale

No recommendation. Unresolved issue. Practices for which insufficient evidence or no consensus regarding efficacy exist

This classification has been used in Tables 1 and 2.

Table 1.Recommendations for hand hygiene (HCWs healthcare workers)a

Recommendation Level of evidence

When hands are visibly dirty or contaminated with proteinaceous IA material or are visibly soiled with blood or other body fluids,

wash hands with either a non-antimicrobial soap and water or an antimicrobial soap and water

If hands are not visibly soiled, use an alcohol-based hand rub IC for routinely decontaminating hands in all other clinical situations

described in items

Decontaminate hands after contact with a patient's intact skin IB (e.g., when taking a pulse or blood pressure, and lifting a patient)

Decontaminate hands after contact with body fluids or excretions, IA mucous membranes, non-intact skin, and wound dressings if hands

are not visibly soiled

Decontaminate hands after removing gloves IB

Before eating and after using a restroom, wash hands with IB a non-antimicrobial soap and water or with an antimicrobial

soap and water

No recommendation can be made regarding the routine use of Unresolved issue non-alcohol-based hand rubs for hand hygiene in healthcare settings

When decontaminating hands with an alcohol-based hand rub, IB apply product to palm of one hand and rub hands together, covering

all surfaces of hands and fingers, until hands are dry

When washing hands with soap and water, wet hands first with water, IB apply an amount of product recommended by the manufacturer to

hands, and rub hands together vigorously for at least 15 seconds, covering all surfaces of the hands and fingers. Rinse hands with water and dry thoroughly with a disposable towel. Use towel to turn off the faucet

Provide personnel with efficacious hand-hygiene products that have IB low irritancy potential, particularly when these products are used

many times per shift

This recommendation applies to products used for hand antisepsis before and after patient care in clinical areas and to products used for surgical hand antisepsis by surgical personnel

➝ Cont.

(4)

Recommendation Level of evidence Do not add soap to a partially empty soap dispenser. This practice IA of "topping up' dispensers can lead to bacterial contamination of soap

Provide HCWs with hand lotions or creams to minimize the occurrence IA of irritant contact dermatitis associated with hand antisepsis or handwashing As part of a multidisciplinary program to improve hand hygiene adherence, IA provide HCWs with a readily accessible alcohol-based hand-rub product To improve hand hygiene adherence among personnel who work in IA areas in which high workloads and high intensity of patient care are

anticipated, make an alcohol-based hand rub available at the entrance to the patients' room or at the bedside, in other convenient locations, and in individual pocket-sized containers to be carried by HCWs For a more detailed description see [10, 11, 13]

Table 2.Recommendations for protective clothing and care of equipment and environment^a

Recommendation Level of evidence

Select protective equipment on the basis of an assessment II of the risk of transmission of micro-organisms

Gloves IB

Wear gloves (clean, non-sterile gloves are adequate) when touching blood, body fluids, secretions, excretions, and contaminated items.

Put on clean gloves just before touching mucous membranes and non-intact skin. Change gloves between tasks and procedures on the same patient after contact with material that may contain a high concentration of micro-organisms. Remove gloves promptly after use, before touching non-contaminated items and environmental

surfaces, and before going to another patient, and wash hands immediately to avoid transfer of micro-organisms to other patients or environments

Mask and eye protection IB

Wear a mask or a face shield to protect mucous membranes of the eyes, nose, and mouth during procedures and patient-care activities that are likely to generate splashes or sprays of blood, body fluids, secretions, and excretions

Gown IB

Wear a gown (a clean, non-sterile gown is adequate) to protect skin and to prevent soiling of clothing during procedures and patient-care activities that are likely to generate splashes or sprays of blood, body fluids, secretions, or excretions. Select a gown that is appropriate for the activity and amount of fluid likely to be encountered.

Remove a soiled gown as promptly as possible, and wash hands to avoid transfer of micro-organisms to other patients or environments

➝ Table 1 Cont.

➝ Cont.

(5)

The Five Main Infection Control Maneuvers to Control Transmission

Hand Hygiene

Hand washing is often referred to as the single most important means of preventing the transmission of healthcare-associated infections (Table 1) [13, 20].

Several studies have shown that the transmission of PPMs from one patient to another via the hands of healthcare workers can result in adverse patient out- comes [21, 22]. The EPIC guidelines based on a systematic review of the available literature suggest that effective hand decontamination can significantly reduce infection rates in high-risk areas such as the ICU. Although there is no level one evidence, i.e., randomized controlled trials, and the ethical approval for such a study would be non-existent, the evidence is based on expert consensus opinion and several observational epidemiological studies [9, 23, 24].

Therefore, hands must be decontaminated and dried thoroughly immediately before each direct patient contact/care episode and after any activity or contact that can result in hands becoming contaminated.

➝ Table 2. cont.

Isolation IB

Place a patient who contaminates the environment or who does not (or cannot be expected to) assist in maintaining appropriate hygiene or environmental control in a private room. If a private room is not available, consult with infection control professionals regarding patient placement or other alternatives

Patient-care equipment IB

Handle used equipment soiled with blood, body fluids, secretions, and excretions in a manner that prevents skin and mucous membrane exposures, contamination of clothing, and transfer of micro-organisms to other patients and environments. Ensure that reusable equipment is not used for the care of another patient until it has been cleaned and reprocessed appropriately. Ensure that single-use items are discarded properly

Environmental control IB

Ensure that the hospital has adequate procedures for the routine care, cleaning, and disinfecting of environmental surfaces, beds, bedrails, bedside equipment, and other frequently touched surfaces, and ensure that these procedures are being followed

Handle, transport, and process used linen soiled with blood, body IB fluids, secretions, and excretions in a manner that prevents skin and

mucous membrane exposures and contamination of clothing, and that avoids transfer of micro-organisms to other patients and environments For a more detailed description see [10–12]

(6)

There are various studies that look at differing products for hand decontamination. Some studies suggest that soap and water is as effective as handwashing products containing antimicrobial agents for decontaminating hands and removing transient micro-organisms [24]. Antimicrobial liquid soap preparations will reduce transient micro-organisms and resident flora and result in hand asepsis. Alcohol-based hand rubs alone are not effective in removing physical dirt or soiling, but will result in substantial reductions in transient micro-organisms. A recent prospective, randomized clinical trial with crossover design, paired data, and blind evaluation in a Spanish ICU demonstrated that alcoholic solutions alone were effective in reducing the number of colony forming units (CFU) on hands after hand washing by 88.2% compared with 49.6% after soap and water [25].

Overall the EPIC project review failed to find compelling evidence for the general use of antimicrobial agents over soap or one antimicrobial agent over another. From the available evidence, based on a review of expert opinion, the authors suggest that acceptability of agents and hand hygiene techniques are the most essential factors for the selection of the best products and compliance with handwashing.

Hand decontamination is, however, often poorly performed and several studies demonstrate that compliance is sub-optimal [26, 27]. The various rea- sons suggested for this include availability of hand decontamination facilities and harsh hand care products causing skin damage. ICUs should therefore have enough sinks with easy access to wash hands. The increased availability of alcohol gel hand rubs, which contain emollients, has also been found to increase compliance [28]. Regular hand decontamination training sessions/audits with compliance could also help increase awareness [29, 30].

However, it must also be appreciated that the saliva or feces of critically ill patients contain high concentrations of PPMs (>10⁸CFU/g of feces), and after contact with such a patient hand contamination can often exceed 10⁵cfu/cm²of finger surface area. Hand decontamination with 0.5% chlorhexidine in 70%

alcohol effectively clears micro-organisms from the hands, but only if the contamination is <10⁴ CFU [31]. Therefore, hand decontamination can only be expected to reduce transmission and not eradicate it completely.

Gloves, Gowns and Aprons (Personal Protective Equipment) Some healthcare-associated infections may easily be transferable between patients by direct transmission via hands of HCW [32] or via inanimate objects such as equipment. The most common PPMs are vancomycin-resistant entero- cocci (VRE), methicillin-resistant Staphylococcus aureus (MRSA), Clostridium difficile, and (often resistant) aerobic Gram-negative bacilli (AGNB). The ICU may often not know the carrier state of a patient on admission and it is imper-

(7)

ative that standard (formerly referred to as universal infection control blood and body fluid) precautions are practised by all HCW. Personal protective equipment should be easily accessible to all HCW in the ICU to ensure compliance with standard precautions [33] and to comply with Health and Safety regulations such as the Personal Protective Equipment Regulations (1994), which are compulsory in some countries, e.g., United Kingdom [34].

Gloves and aprons should be used routinely when handling blood and/or body fluids from all patients to protect the HCW and prevent transmission to other susceptible individuals. These measures ensure all patients are treated equally without breaching confidentiality.

Latex gloves are one of the most effective barriers against micro-organisms [32], although several studies have shown poor compliance [26, 27], with gloves not always being worn when required, or conversely being worn for prolonged periods and hands not being washed following removal, leading to potential hand contamination. Gloves can also cause adverse reactions and skin sensitiv- ity in both the HCW and patients [32] and their use should not be indiscrimi- nate. Risk assessments should be undertaken to establish when to use gloves and which particular type of glove to use [32]. Sterile gloves should be worn for all invasive procedures, contact with sterile sites, and non-intact skin. Non-sterile gloves are suitable for all procedures, which involve contact with mucous membranes, and where a risk assessment has indicated exposure to blood or body fluids, secretions, and excretions. Gloves should also be single-use items and discarded after each activity as clinical waste, followed by hand decontamination.

Disposable plastic aprons are just as effective as gowns in protecting the clothing of HCW. Many studies advocate gown usage to prevent the transmis- sion of PPMs, in particular C. difficile and VRE [26, 35, 36]. However, although many of the studies demonstrated a reduction in the incidence of infection due to the above PPMs, it was difficult to differentiate between individual factors, including severity of illness, type of underlying disease, and the use of antibiotics. Antibiotics that respect the patient’s normal flora, i.e., ecology, have been shown to control VRE and C. difficile, [37, 38]. Furthermore, a systematic review undertaken by the EPIC project authors identified two randomized controlled trials where gown usage in special care baby units failed to reduce infection rates [39, 40]. However, where there is a possibility of contamination of the clothing of HCW with blood or body fluids, it is recommended that disposable plastic aprons be worn. Gowns are only required where there is a risk of gross contamination or splashing, such as in major burn patients or severe trauma, and in this event the gowns should be made of a fluid-repellent material.

In the ICU or emergency department the HCW should also have access to facial or eye protection. Although facemasks have not been found to be benefi- cial, personal respiratory protection is required for patients with tuberculosis

(8)

[41]. The masks should be specialized respiratory masks developed solely for this purpose, e.g., RP4 masks. Potential splashing of blood and body fluids to the face or eyes has also been reported in ICU staff, particularly following suc- tion of endotracheal tubes. Specially developed eye protection and visors should be worn if a risk assessment indicates that this is a likely occurrence. If such an exposure to blood and body fluids were to happen, the HCW must report this as an inoculation incident and be treated accordingly.

Isolation

In addition to standard infection control precautions, procedures may have to be implemented if a patient is identified as carrying or being infected with certain transmissible or antibiotic-resistant PPMs. PPMs such as those transmitted via respiratory droplets fall into this category. The patient may require nursing in a side room with isolation facilities. Several studies have demonstrated that isolation barrier precautions are effective in achieving a significant reduction in transmission, particularly in relation to VRE, C. difficile, and res- piratory syncytial virus (RSV) and in some outbreaks of MRSA and AGNB [42–47]. However, it may be difficult to isolate patients due to the availability of such facilities and the increasing numbers of patients with antibiotic-resistant PPMs that may require isolation. In this event, a risk assessment should be undertaken to decide the mode and likelihood of transmission occurring in other susceptible patients on the ICU. The patient can then be isolated on the basis of greatest risk, or patients with the same microbiologically confirmed pathogen may be nursed together, i.e., cohort nursing in the same area. For example, infants with bronchiolitis caused by RSV may often have to be cared for in areas with cohort facilities in the midst of the annual winter epidemic [48].

Isolation barrier precautions should be implemented when there is a patient on the ICU suspected of having a resistant PPM. Furthermore, these precautions should be commenced on suspicion without awaiting microbiological confirmation, e.g., C. difficile or RSV in a high-risk environment such as the ICU. However, precautions should be simple to follow and implement to ensure compliance, avoiding unnecessary time-consuming additional precautions, which could increase the workload of the ICU staff. Such precautions could also increase psychological effects for the ICU patient [48].

Role of Patient-Care Equipment

Patient-care equipment has been implicated in several outbreaks of infection in ICU patients [43, 47, 49], in particular where there is movement of equipment

(9)

between patients. Common procedure and device-related infections are dealt with in Chapter 11. However, all equipment used for patients on the ICU can be divided into high-, intermediate-,low- and minimal risk items.

High-risk items include any device that comes into contact with a break in the skin or mucous membrane or is introduced into a sterile body area, e.g., surgical instruments, dressings, catheters, and parenteral fluids. These items should be sterile and suitable for sterilization.

Intermediate-risk items include devices that come into contact with intact mucous membranes, body fluids, or are contaminated with particularly virulent or readily transmissible organisms, or items to be used on highly susceptible patients or sites, e.g., endoscopes, respiratory equipment. These items require disinfecting.

Low-risk items include those that come into contact with normal or intact skin, e.g., thermometers, stethoscopes, washbowls, toilets, and bedding.

Cleaning and drying is usually sufficient, but disinfecting is required in the case of a patient with a known infection risk, e.g., MRSA.

Minimal-risk items include those not in contact with the patient, such as floors and surfaces. Cleaning and drying is usually adequate unless in an outbreak situation.

Equipment in the ICU is used with the sickest, most susceptible patients.

Therefore, equipment used for invasive procedures should be properly steril- ized after every patient, not only those known to be infected. Patient equipment such as stethoscopes and thermometers should be designated to individual patients wherever possible. When dedicated equipment is not practical, e.g., portable X-ray machines and ultrasound machines, staff should ensure that such equipment is cleaned between patients.

All environments, except those maintained under sterile conditions, can harbor PPMs [48], the ICU is no exception. Ventilation equipment, humidifiers, analyzers, and transducers have all been implicated in outbreaks of environmental external transmission episodes. PPMs commonly involved include Acinetobacter sp. and Pseudomonas aeruginosa and non-aeruginosa (Stenotrophomonas maltophilia and Burkholderia cepacia), MRSA, VRE, and C.

difficile. Although many epidemics of nosocomial infections have stemmed from reservoirs of PPMs in the inanimate hospital environment, the contribu- tion of the environment to the acquisition and spread of endemic nosocomial infections is thought to be in most cases insignificant [50, 51]. PPMs have to be transmitted from the environment to the patient in order to cause an infection.

This is usually by airborne, common vehicle or direct contact. Patient carriers are required for contamination of the environment [51], except for Aspergillus species [52], and these can often be prevented by basic infection control meas- ures directed to the external inanimate source [53].

(10)

Role of Patient-Care Environment

A clean environment is necessary to provide the required background to good standards of hygiene and asepsis and to maintain the confidence of patients and the morale of staff. The ICU environment is categorized as a high-risk area under the Standards for Environmental Cleanliness in Hospitals Guidelines [54]

and recommendations that cleaning equipment and solutions should conform to hospital policy. When considering the role of the environment in the ICU, the design of the unit should ensure that there is adequate space between beds to allow easy access of staff and equipment, preventing overcrowding and enabling environmental cleaning. Furniture and fixtures should be kept to a minimum, and be of materials that are easy to clean [53, 54].

How to Evaluate the Effectiveness of an Infection Control Maneuver: Surveillance

Surveillance has been accepted as a valid method to evaluate the effectiveness of infection control maneuvers on the ICU.

What Is Surveillance?

Surveillance has been defined as “the systematic, active ongoing observation of the occurrence and distribution of disease in a population” [55]. The focal point for infection control activities on the ICU is a system of close observation designed to establish and maintain a database that describes rates of infection due to micro-organisms acquired on the ICU. There are three fundamental ele- ments: (1) data collection; (2) data analysis and interpretation; (3) data report- ing to individuals who are in a position to take appropriate action in order to achieve control. However, it is important to appreciate that the implementation of a surveillance program on its own does not automatically achieve control of infection due to ICU-related micro-organisms.

The Infection Surveillance Program Based on Surveillance of Carriage

Knowledge of the carrier state using surveillance cultures of throat and rectum is essential to distinguish primary endogenous infections due to micro-organisms present in the admission flora from secondary endogenous and exogenous infections due to micro-organisms acquired on the ICU following transmission. Recent studies using surveillance cultures of throat and rectum to detect the carrier state demonstrate that only infections occurring after 1 week of ICU stay are due to

(11)

microbes transmitted via the hands of HCW [56–59]. The incidence varies between 15% and 40%, depending on the severity of illness. PPMs related to the ICU environment are first acquired in the oropharynx. In the critically ill, oropharyngeal acquisition invariably leads to secondary or super carriage. The subse- quent build up to digestive tract overgrowth, defined as ≥10⁵PPMs per milliliter of saliva or gram of feces, which can then result in colonization of normally sterile internal organs, takes a few days. Finally, it is the degree of impaired immunity of the ICU patients that determines the colonization, leading to an established secondary endogenous infection or superinfection. The other type of ICU infection is exogenous which is [60] due to breaches of hygiene. The causative bacteria are also acquired on the unit but are never present in the throat and/or gut flora of patients. For example, long-stay patients, particularly those who receive a tra- cheostomy, on respiratory units are at high risk of exogenous lower airway infections. Purulent lower airway secretions yield a micro-organism that has never been previously carried by the patient in the digestive tract flora, or indeed in their oropharynx. Although both the tracheotomy and the oropharynx are equally accessible for bacterial entry, the tracheotomy tends to be the entry site for bacteria that colonize/infect the lower airways. However, the major infection problem is primary endogenous due to micro-organisms that the patient imports into the ICU in their admission flora. The proportions of primary endogenous infections vary between 60% and 85%, they typically occur within the 1st week of ICU stay, and the PPMs involved do not bear any relationship to the ICU ecology. A recent study compared the traditional 48-h cut-off and the criterion of the carrier state and found that the time cut-off significantly overestimated the magnitude of the nosocomial problem [61]. This approach to the carrier state may be more useful for interhospital comparison, as only infections due to micro-organisms acquired on the different units are compared, independent of illness severity [59].

Infection surveillance systems based on surveillance of the carrier state allows knowledge of the pathogenesis of the infections and consequently the type of preventative measures to be implemented. These include:

1. Short course of systemic antibiotics, i.e., cefotaxime, to prevent primary endogenous infections

2. Hygienic measures, i.e., handwashing and isolation, and non-absorbable oropharyngeal and intestinal antibiotics to prevent secondary endogenous infections

3. Hygiene to prevent exogenous infections.

Additionally, surveillance samples of throat and rectum allow the detection of the carrier state of resistant micro-organisms, e.g., MRSA, on admission permit- ting swift isolation of the patient before the diagnostic samples yield MRSA, con- firming an infection with MRSA. Secondly, in identifying the population with primary endogenous infections, the classification using the carrier state avoids blam- ing staff for all infections after 48 h for which they are not responsible. The knowl-

(12)

edge of the carrier state thus prevents fruitless investigation of apparent cross- infection episodes. Thirdly, without surveillance samples, exogenous infections, which can occur at any time in the ICU due to contaminated equipment, are impossible to recognize, at least at an early stage when only diagnostic samples such as tracheal aspirate, urine, and blood have been tested. Finally, the knowledge of the carrier state allows detection of ongoing transmission and an impending outbreak, as the surveillance cultures become positive for, e.g., a multi-resistant Acinetobacter baumannii at an early stage before the patients develop infection [62] and before culturing hands of carers yields the same strain. Active surveillance for VRE has also been found to significantly prevent further transmission in other studies [63–65]. Therefore, the cost of surveillance sampling is weighed against the cost of a potentially more expensive reactive approach. Although a recent review of the available literature on the subject of surveillance found the use of surveillance cultures controversial [66], this struc- tured surveillance method prevents the often routine diagnostic sampling practice of several ICUs and again can be a more cost-effective approach [67, 68].

Which samples are required for surveillance? Surveillance of both infection and carrier state requires surveillance samples of throat and rectum, besides diagnostic samples of lower airway secretions, urine, wound discharge, and blood.

Surveillance samples are defined as samples obtained from body sites where PPMs are carried, i.e., the digestive tract comprising the oropharyngeal cavity and rectum [69]. Surveillance swabs are different from surface and diagnostic samples.

Surface samples are swabs from the skin such as axilla, groin, umbilicus, nose, eye, and ear. These sites should not form part of the surveillance sampling proto- col, as positive surface swabs merely reflect oropharyngeal and rectal carrier state.

Diagnostic samples, i.e., from sites that are normally sterile, such as lower airways, blood, bladder, and from skin lesions, should only be obtained when clinically indicated in order to establish a microbiological cause for the clinical diagnosis of inflammation, both generalized and/or local.

The aim of obtaining surveillance cultures, in both a qualitative and (semi)quantitative way, is the determination of the microbiological endpoint of the carrier state of PPMs (qualitative) and overgrowth (quantitative). Carriage or carrier state exists when the same micro-organism is isolated from at least two consecutive surveillance samples obtained from the ICU patient in any concentration over a period of at least 1 week. Carriage indicates persistence of micro-organisms and is distinguished from acquisition or transient presence.

Overgrowth, defined as ≥10⁵ PPMs per milliliter of saliva and/or gram of feces, is a risk factor for developing infection. In a population of 85 patients who carried MRSA in the digestive tract, the risk of developing at least one infection was 64% when overgrowth existed. However when MRSA carriage existed without overgrowth the risk was 12.5% [68].

(13)

Which patients may benefit from carriage surveillance? Most infection surveil- lance programs include all patients admitted to the ICU whether they stay a few days or 2 weeks [70, 71]. The inclusion of a large number of relatively short- stay patients with a low risk of infection tends to dilute the total rates of infection by increasing the size of the denominator. However, low percentages may look good to managers, but do not allow room for improvement, i.e., the detection of a significant reduction in infection rate following the introduction of an intervention [58]. We believe that all patients requiring mechanical ventilation for a minimum of 3 days should be part of an ongoing surveillance programme of both infection and carriage.

The role of surveillance samples in monitoring the efficacy of selective digestive decontamination (SDD) is a debatable issue. The reduction in the incidence of mortality and pneumonia has been shown when SDD has been admin- istered with or without taking surveillance samples. Thus, it may be acceptable to use SDD even when the microbiology laboratory does not process the surveillance samples. We recommend obtaining surveillance samples to monitor the efficacy of SDD in order to detect failures due to inappropriate administra- tion, ileus, and low compliance.

Finally, it is generally accepted that surveillance samples should be taken from patients at risk of acquiring PPMs during outbreaks or endemicity of resistant PPMs. The magnitude of the problem of resistant PPMs can be identified only by detecting carriers. The diagnostic samples represent “the tip of the iceberg” (Fig. 1).

0 2 4 6 8 10 12 14 16

Jan July Jan July Jan July

ICU-ACQUIRED IMPORTED MRSA incidence : diagnostic samples New cases

Months

2000 2001 2002

0 2 4 6 8 10 12 14 16

Jan July Jan July Jan July

ICU-ACQUIRED IMPORTED MRSA incidence : surveillance and diagnostic samples New cases

Months

2000 2001 2002

A

B

Fig. 1. A The monthly incidence of new cases of methicillin-resistant Staphylococcus aureus (MRSA) detected by diagnostic samples in an intensive care unit (ICU). B The inci- dence of new cases of MRSA using surveillance and diagnostic samples during the same period in the same ICU [68]

(14)

The Minimum Infection Database

The traditional infection surveillance systems are based on the recommendations of the CDC [71, 72]. The aim of this approach is to provide a simple and efficient tool:

1) to define the rates of the most relevant ICU-acquired infections, i.e., pneumonia related to mechanical ventilation, bloodstream infections related to intravascular devices, and urinary tract infections related to bladder catheters

2) to monitor the trend of these rates in the ICU

3) to assess the impact of any preventive measure (Fig. 2).

The infection indicators recommended by the CDC are based on the relative weight of the following criteria: (1) clear case definition; (2) ease of surveillance system, importance of the event; (3) potential of intervention to reduce rates;

(4) availability of denominator device-days; (5) ease of collection of the denominator device-days [73].

The selected indicators include:

Ventilator-associated pneumonia rate=number of ventilator-associated pneumonia/number of ventilator-days

Central line associated bloodstream infection rate=central line-associated bloodstream infections/number of central line-days

Fig. 2. The incidence of ventilator-associated pneumonia expressed as the number of pneumonias divided by 1,000 ventilation-days significantly decreased following imple- mentation of selective digestive decontamination (SDD) form 12.4 to 3.6 (P<0.001) [80]

Ventilator-associated pneumonia rates : before and after implementing SDD

Months Rates

1997 1998 1999 2000

SDD

Ventilator-associated pneumonia rates : before and after implementing SDD

0 5 10 15 20 25 30

Jan July Jan July Jan July Jan July _Months

Rates

1997 1998 1999 2000

SDD

(15)

Urinary catheter-associated urinary tract infection rate=number of urinary catheter-associated urinary tract infection/number of urinary catheter-days.

This minimum data set has been proposed for use in the ICU, even in those with limited resources, because of the low workload for data acquisition and the relevance of the information retrieved [74–76].

Nevertheless the use of diagnostic samples may only underestimate the level of transmission of PPMs because diagnostic samples are generally taken on clinical indication for microbiological confirmation. Moreover, diagnostic samples do not detect transmission [62, 68].

The strengths and the weaknesses of both infection surveillance systems, based on surveillance samples and diagnostic samples, are summarized in Table 3.

Strengths Surveillance of infection (solely diagnostic samples)

• Already routine

• Easy to fulfil

• Number of infections per 1,000 device- days: useful to know the trends in infection rates in one unit

Surveillance of infection/carriage (diagnostic samples combined with

surveillance samples)

• More accurate estimation of infections due to ICU-acquired micro-organisms

• Early implementation of the appropriate preventive measures according to the pathogenesis of the infections

• Detection of resistance at an early stage

• Detection of transmission at an early stage

• Indispensable in control of an outbreak

• Monitoring the efficacy of selective digestive decontamination

Weaknesses Surveillance of infection (solely diagnostic samples)

• Substantial delay between the detection of a problem and the implementation of the appropriate measures to control it, because of the extra work required to identify the pathogenesis of the problem

• Cost-effectiveness: has to be tested

• Time cut-off of 48 h: not accurate for the estimation of infection due to ICU micro- organisms

• Value of method for interhospital comparison: limited

• Detection of resistance, transmission and outbreaks: late

Surveillance of infection/carriage (diagnostic samples combined with

surveillance samples)

• Workload for laboratory is higher

• Cost-effectiveness: has to be tested

• Value of method for interhospital comparison: has to be tested

• Surveillance cultures: unpopular amongst traditional microbiologists

Table 3.Strengths and weaknesses of both surveillance methods of infection only and of infection combined with carriage

(16)

How to Implement an Infection Control Program

Implementing an effective evidence-based infection control program in the ICU not only relies on establishing and reviewing the best available evidence, but it is essential that the evidence is practical and easy to utilize in order to achieve compliance with the recommended guidelines and policies.

1. The intensivist together with an ICU nurse interested in infection control, in general, take the initiative to design an infection control program. They invite the infection control team of the hospital, the clinical microbiologist, and pharmacist to join the ICU infection control team. In some cases the hospital infection control team approaches ICU staff to develop the program.

2. The resultant ICU infection control team defines the content of the infection control program that should initially include:

(a) How to implement the five infection control maneuvers (hand hygiene, protective clothing, isolation, care of the equipment, and care of the environment) in order to control the transmission of PPMs

(b) The definition of the population who may benefit from SDD and how to implement it [77, 78]

(c) The minimum data set to evaluate the quality of infection control (Fig. 2) (d) The surveillance cultures (throat and rectum) targeting carriage of resistant PPMs, i.e., MRSA, P. aeruginosa, Acinetobacter spp. (Fig. 1) (e) How to report the results of the database and liaize with HCW.

Regular meetings of the ICU infection control team are indispensable to dis- cuss current patient-related problems and relevant infection control issues.

3. Regular surveillance samples to monitor carriage of patients who receive SDD in order:

(a) To identify the preventative measure to be implemented

(b) To monitor the efficacy of SDD in eradicating the abnormal carrier state (Fig. 3)

4. The antibiotic policy and surveillance of antimicrobial usage are discussed in Chapters 12 and 28.

(17)

References

1. Waterer GW, Wunderink RG (2001) Increasing threat of Gram-negative bacteria. Crit Care Med 4 [Suppl]:N75–N81

2. Warren DK, Fraser VJ (2001) Infection control measures to limit antimicrobial resistance. Crit Care Med 29:N128–N134

3. Fridkin SK (2001) Increasing prevalence of antimicrobial resistance in intensive care units. Crit Care Med 4 [Suppl]:N64–N68

4. Masterton RG, Teare EL (2001) Clinical governance and infection control in the United Kingdom. J Hosp Infect 47:25–31

5. NHS Executive (1999) Health Service Circular 1999/132 Governance in the New NHS 21 May. Department of Health, London

6. Condon RE, Haley RW, Lee JT et al (1988) Does infection control control infection?

Arch Surg 123:250–256

7. Wenzel RP (ed) (1993) Prevention and control of nosocomial infections, 2nd edn.

Williams and Wilkins, Baltimore, USA

Fig. 3.The incidence density of secondary endogenous (_■) and exogenous infections (∆) due to newly acquired micro-organisms that were not present in the admission flora.

Micro-organisms causing exogenous infections were not cultured from the surveillance samples of throat and rectum. Secondary endogenous infections were due to micro-organisms that had been through a digestive tract phase. The micro-organisms were acquired on the pediatric intensive care unit in both types of infection. Two thresholds were used for exogenous and secondary endogenous infections, respectively. The thresholds were calcu- lated at the 95 % confidence interval from the first 24 months of data, assuming the infection rate data conforms to the Poisson distribution. Exogenous infections rose above the threshold on twenty occasions whereas secondary endogenous infections were above the threshold three times (from Sarginson RE, Taylor N, Reilly N, Baines PB, Van Saene HKF (2004) Infection in prolonged pediatric critical illness: A prospective four-year study based on knowledge of the carrier state. Crit Care Med 32:839-847, with permission) [59]

0 1 2 3 4 5 6

Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun Sep Dec

Bacterial/yeast infections/month of 100 patient days

1999 2000 2001 2002

(18)

8. Ayliffe GAJ, Fraise A, Geddes A, Mitchell K (1998) Control of hospital infection; a practical handbook, 4th edn. Arnold, London

9. Ward V, Wilson J, Taylor L et al (1997) Preventing hospital-acquired infection: clinical guidelines, vol 42. Public Health Laboratory Service, London

10. Shojania KG, Duncan BW, McDonald KM et al (2001) Making health care safer: a critical analysis of patient safety practices. Evidence Report/Technology Assessment no.

43 (Prepared by the University of California at San Francisco—Stanford Evidence- based Practice Center under contract no. 290-97-0013). AHRQ Publications no. 01- E058, Rockville, Md.

11. Pratt RJ, Pellowe C, Loveday HP et al (2001) The EPIC Project: developing national evidence-based guidelines for preventing healthcare associated infections. J Hosp Infect 47 [Suppl]:1–34

12. Garner JS and the Hospital Infection Control Practices Advisory Committee.

Guideline for Isolation Precautions in Hospitals. http://www.cdc.gov/ncidod/hip/isolat/isolat.htm

13. Boyce JM, Didier Pittet D (2002) Guideline for hand hygiene in health-care settings.

Recommendations of the healthcare infection control practices advisory committee and the hicpac/shea/apic/idsa hand hygiene task force. Am J Infect Control 30:S1–S46 14. Sackett LD, Rosenberg W, Haynes BR (2000) Evidence-based medicine: how to practi-

ce and teach EBM, 2nd edn. Churchill Livingstone, Edinburgh

15. Harbour R, Miller J (2001) Education and debate: a new system for grading recommendations in evidence based guidelines. BMJ 323:334–336

16. Ayliffe GAJ (2000) Evidence-based practises in infection control. Br J Infect Control 1:4:5–9

17. Jenner EA, Mackintosh C, Scott GM (1999) Infection control–evidence into practice. J Hosp Infect 42:91–104

18. Ward D (2000) Implementing evidence-based practice in infection control. Br J Nurs 9:267–271

19. NHS Executive (1996) Clinical guidelines. Department of Health, London

20. Infection Control Nurses Association (1998) Guidelines for hand hygiene. ICNA/Deb, West Lothian, p 13

21. Simmons B, Bryant J, Neiman K et al (1990) The role of handwashing in the prevention of endemic intensive care unit infections. Infect Control Hosp Epidemiol 11:589–594

22. Sproat LJ, Inglis TJJ (1994) A multicentre survey of hand hygiene practice in intensive care units. J Hosp Infect 26:137–148

23. Gould D (1991) Nurses’ hands as vectors of hospital-acquired infection: a review. J Adv Nurs 16:1216–1225

24. Reybrouck G (1983) Role of the hands in the spread of nosocomial infections. J Hosp Infect 4:103-110

25. Zaragoza M, Salles M, Gomez J et al (1999) Handwashing with soap or alcoholic solutions? A randomized clinical trial of its effectiveness. Am J Infect Control 27:258–261 26. Larson E, Kretzer EK (1995) Compliance with handwashing and barrier precautions. J

Hosp Infect 30 [Suppl]:88–106

27. Zimakoff J, Stormark M, Oleson Larson S (1993) Use of gloves and handwashing beha- viour among healthcare workers in intensive care units: a muticentre investigation in four hospitals in Denmark and Norway. J Hosp Infect 24:63–67

28. Pittet D (2000) Improving compliance with hand hygiene in hospitals. Infect Control Hosp Epidemiol 6:381–386

29. Pittet D, Hugonnet S, Harbarth S et al (2000) Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 356:1307–1312

(19)

30. Earl ML, Jackson MM, Rickman LS (2001) Improved rates of compliance with hand antisepsis guidelines: a three-phase observational study. Am J Nurs 101:26–33 31. Murray AE, Chambers JJ, van Saene HKF (1998) Infections in patients requiring ven-

tilation in intensive care: applications of a new classification. Clin Microbiol Infect 4:94–102

32. Infection Control Nurses Association (1999) Glove usage guidelines. ICNA and Regent Medical, London, p 43

33. Expert Advisory Group on AIDS and the Advisory Group on Hepatitis (1998) Guidance for clinical health care workers: protection against infection with blood- borne viruses. Department of Health, London, p 46

34. Health and Safety Executive (1992) Personal protective equipment regulation.

Guidance on regulations. HMSO, London

35. Slaughter S, Hayden MK, Nathan et al (1996) A comparison of the effect of universal use of gloves and gowns with that of glove use alone on acquisition of vancomycin- resistant enterococci in a medical intensive care unit. Ann Intern Med 125:448–456 36. Kenny H, Larson E (2000) The efficacy of cotton cover gowns in reducing infection in

nursing neutropenic patients: an evidence-based study. Int J Nurs Prac 6:3:135–139 37. Donskey CJ, Tanvir K, Chowdhry et al (2000) Effect of antibiotic therapy on the den-

sity of vancomycin-resistant enterococci in the stool of colonized patients. N Engl J Med 343:1925–1932

38. Borriello SP (1990) The influence of the normal flora on Clostidium difficile colonisa- tion of the gut. Ann Med 22:61–67

39. Rush J, Fiorino-Chiovotti R, Kaufman K et al (1990) A randomized controlled trial of a nursery ritual: wearing cover gowns to care for healthy newborns. Birth 17:25–30 40. Birenbaum EJ, Gloriso L, Rosenburger C et al (1990) Gowning on a postpartum ward

fails to decrease colonisation in the newborn infant. Am J Dis Child 144:1031–1033 41. The Interdepartmental Working Group on Tuberculosis (1998) The prevention and

control of tuberculosis in the United Kingdom: UK Guidance on the prevention and control of transmission of 1. HIV-related tuberculosis 2 drug-resistant, including mul- tiple drug-resistant, tuberculosis. Department of Health, London, p 94

42. Preston GA, Larson EL, Stamm WE (1981) The effect of private isolation rooms on patient care practices, colonization and infection in an intensive care unit. Am J Med 70:641–645

43. Bonten MJM, Hayden MK, Nathan C et al (1996) Epidemiology of colonisation of patients and environment with vancomycin-resistant enterococci. Lancet 348:1615–1619

44. Hanna H, Raad I, Gonzalez V et al (2000) Control of nosocomial Clostridium difficile transmission in bone marrow transplant patients. Infect Control Hosp Epidemiol 21:226–228

45. Madge P, Payton JY, McColl et al (1992) Prospective controlled study of four infection control procedures to prevent nosocomial infection with respiratory syncytial virus.

Lancet 340:1079–1083

46. Girou E, Pujade G, Legrand P, Cizeau F, Brun-Buisson C (1998) Selective screening of carriers for control of methicillin-resistant Staphylococcus aureus (MRSA) in high- risk hospital areas with a high level of endemic MRSA. Clin Infect Dis 27:543–550 47. Fierobe L, Lucet JC, Decre D, Muller-Serieys C, Deleuze A, Joly-Guillou ML, Mantz J,

Desmonts JM (2001) An outbreak of imipenem-resistant Acinetobacter baumannii in critically ill surgical patients. Infect Control Hosp Epidemiol 22:35–40

48. Kirkland K, Weinstein JM (1999) Adverse effects of contact isolation. Lancet 354:1177–1178

49. Jones JS, Hoerle D, Riekse R (1995) Stethoscopes: a potential vector of infection? Ann

(20)

Emerg Med 26:296–299

50. NHS Estates (2001) Infection control in the built environment: design and planning.

Department of Health, HMSO, London

51. van Saene HKF, van Putte JC, van Saene JJM et al (1989) Skin flora in a long stay hospital is determined by the patient’s oral and rectal flora. Epidemiol Infect 102:231–238 52. Manuel RJ, Kibbler CG (1998) The epidemiology and prevention of invasive

Aspergillus infection. J Hosp Infect 39:95–109

53. O’Connell NH, Humphreys H (2000) Intensive care unit design and environmental factors in the acquisition of infection. J Hosp Infect 45:255–262

54. NHS Estates (2001) National standards of cleanliness for the NHS. Department of Health, HMSO, London

55. Hughes JM (1987) Nosocomial infection surveillance in the United States: historical perspective. Infect Control 8:450–453

56. Silvestri L, Monti Bragadin C, Milanese et al (1999) Are most ICU infections really nosocomial? A prospective observational cohort study in mechanically ventilated patients. J Hosp Infect 42:125–133

57. De la Cal MA, Cerda E, Garcia-Herrio P et al (2001) Pneumonia in patients with severe burns. A classification according to the concept of the carrier state. Chest 119:1160–1165

58. Petros AJ, O’Connell, Roberts C et al (2001) Systematic antibiotics fail to clear multi- drug-resistant Klebsiella from a pediatric ICU. Chest 119:862–866

59. Sarginson RE, Taylor N, Reilly N et al (2004) Infection in prolonged pediatric critical illness: a prospective four year study based on knowledge of the carrier state. Crit Care Med 32:839-847

60. Morar P, Singh V, Jones AS et al (1998) Impact of tracheotomy on colonization and infection of lower airways in children requiring long-term ventilation. Chest 113:77–85

61. Silvestri L, Sarginson RE, Hughes J et al (2002) Most nosocomial pneumonias are not due to nosocomial bacteria in ventilated patients. Evaluation of the accuracy of the 48h time cut-off using carriage as gold standard. Anaesth Intensive Care 30:275–282 62. Chetchotisakd P, Phelos CL, Hartstein AI (1994) Assessment of bacterial cross tran-

smission as a cause of infections in patients in intensive care units. Clin Infect Dis 18:929–937

63. Siddiqui AH, Harris AD, Hebden J et al (2002) The effect of active surveillance for vancomycin-resistant enterococci in high-risk units on vancomycin-resistant enterococci incidence hospital-wide. Am J Infect Control 30:40–43

64. Zuckerman RA, Steele L, Venezia RA et al (1999) Undetected vancomycin-resistant Enterococcus in surgical intensive care unit patients. Infect Control Hosp Epidemiol 20:685–686

65. Hendrix CW, Hammond JMJ, Swoboda SM et al (2001) Surveillance strategies and impact of vancomycin-resistant enterococcal colonization and infection in critically ill patients. Ann Surg 233:259–265

66. Glupczynski Y (2001) Usefulness of bacteriological surveillance cultures for monitoring infection in hospitalised patients: a critical appraisal. Acta Clin Belg 56:38–45 67. Langer M, Corretto E, Haeusler EA (2001) Infection control in ICU: back (forward) to

surveillance samples? Intensive Care Med 27:1561–1563

68. de la Cal MA, Cerdá E, van Saene HKF, García-Hierro P, Negro E, Parra ML, Arias S, Ballesteros D (2004) Effectiveness and safety of enteral vancomycin to control ende- micity of methicillin-resistant Staphylococcus aureus in a medical/surgical intensive care unit. J Hosp Infect 56:175-183

(21)

69. van Saene HKF, Taylor N, Reilly N et al (2001) The usefulness of surveillance cultures:

a prospective cohort study on the ICU. In: van Saene HKF, Sganga G, Sivestri L (eds) Topics in anaesthesia and critical care. Infection in the critically ill: an ongoing chal- lenge. Springer Verlag Italia, Milan, pp 58–80

70. Kollef MH, Sherman G, Ward S et al (1999) Inadequate antimicrobial treatment of infections. Chest 115:462–464

71. Gaynes RP, Horan TC (1999) Surveillance of nosocomial infections. In: Mayhall CG (ed) Hospital epidemiology and infection control, 3rd edn. Lippincott Williams and Wilkins, Philadelphia, pp 1285–1318

72. Horan TC, Emori TG (1997) Definitions of key terms used in the NNIS System. Am J Infect Control 25:112–116

73. The Quality Indicator Study Group (1995) An approach to the evaluation of quality indicators of the outcome of care in hospitalized patients, with a focus on nosocomial infection indicators. Am J Infect Control 23:215–222

74. Suetens C, Savey A, Labeeuw J, Morales I (2002) The ICU-HELICS programme:

towards European surveillance of hospital-acquired infections in intensive care units.

Eur Surveill 7:127–128

75. de la Cal MA, Cerda E (1997) Surveillance and control of infections in the intensive care unit: rates, resistance, and carrier state. Enferm Infecc Microbiol Clin 15 [Suppl 3]:47–52

76. Álvarez-Lerma F, Palomar M, Olaechea P, De la Cal MA, Insausti J, Bermejo B, y Grupo de Estudio de Vigilancia de Infección Nosocomial en UCI (2002) Estudio nacional de vigilancia en unidades de cuidados intensivos. Informe del año 2000. Med Intensiva 2:39–50

77. Krueger WA, Lenhart FP, Neeser G et al (2002) Influence of combined intravenous and topical antibiotic prophylaxis on the incidence of infections, organ dysfunctions, and mortality in critically ill surgical patients: a prospective, stratified, randomized, dou- ble blind, placebo-controlled clinical trial. Am J Respir Crit Care Med 166:1029-1037 78. de Jonge E, Schultz MJ, Spanjaard L et al (2003) Effects of selective decontamination of

the digestive tract on mortality and acquisition of resistant bacteria in intensive care:

a randomised controlled trial. Lancet 362:1011-1016

79. Liberati A, D’Amico R, Pifferi S et al (2004) Antibiotic prophylaxis to reduce respiratory tract infections and mortality in adults receiving intensive care [Cochrane Review]. In: The Cochrane Library. Issue 1, Chichester, UK: John Wiley & Sons Ltd 80. Parra ML, Arias S, de la Cal MA, Frutos F, Cerdá E, García-Hierro P, Negro E (2002)

Descontaminación selectiva del tubo digestivo: efecto sobre la incidencia de la infec- ción nosocomial y de los microorganismos multirresistentes en enfermos ingresados en unidades de cuidados intensivos. Med Clin (Barc) 118:361–364